Ride Comfort Sensitivity Analysis Of Portal Axle Using Adams/Car

Yorumlar

Transkript

Ride Comfort Sensitivity Analysis Of Portal Axle Using Adams/Car
Ride Comfort Sensitivity Analysis Of Portal Axle Using Adams/Car
Salih Deniz KIZILCA, Emre SERT, Sertaç DİLEROĞLU
Anadolu Isuzu Otomotiv A.Ş
Abstract
Ride comfort had been developed by satisfying many requirements from engineering design
guides, CAE (computer aided engineering) and vehicle testing. So that, this work is an
investigation on the ride behavior of the portal axle. Test results showed that damper and air spring
geometries cause the bounce motion which affects the passengers' sensation of discomfort.
Therefore, vehicle has been modelled and developed ride comfort using Adams/Car. The studies
show that the damper and air spring attachment points are one of the sensitive parameter to affect
the ride comfort characteristics. Therefore, damper and air spring “x” and “y” attachment points
have been selected as a design variable.
Keywords: Portal Axle, Parameter Optimization, Suspension Analysis, Adams/Car, Low Floor
Bus
Özet
Sürüş konforu, CAE (bilgisayar destekli mühendislik), araç testleri ve mühendislik tasarım
kılavuzları ihtiyaçlarına cevap verecek şekilde geliştirilmiştir. Bu nedenle, bu çalışmada, portal
aksın sürüş davranışı üzerine etkileri incelenmiştir. Test sonuçları, amortisör ve havalı yay
geometrilerinin, yolcuların konforsuz hissedeceği rahatsız sıçrama hareketine neden olduğunu
gösterdiği için Adams/Car kullanılarak aracın sanal modeli oluşturulup sürüş konforu
geliştirilmiştir. Çalışmalar, amortisör ve hava körüğü bağlantı noktalarının sürüş konforu
karakteristikliğini etkileyen duyarlı parametrelerden biri olduğunu gösterdiği için amortsiör ve
hava körüğünün "x" ve "y" bağlantı noktaları dizayn değişkenleri olarak seçilmiştir.
Anahtar Kelimeler: Portal Aks, Parametre Optimizasyonu, Süspansiyon Analizi, Adams/Car,
Alçak Tabanlı Otobüs
1. Introduction
The comfort level of a vehicle is one of the most important factor in a vehicle's subjective
evaluation. Vehicle ride analysis isused to design the suspension to ensure that passenger
comfort. There are two basic types of elements in conventional suspension systems that define
the performance of the suspension. The elements are springs and dampers; the role of the
spring in a vehicle's suspension system is to store the energy and support the static weight of
the vehicle, while the role of the dampers to dissipate vibrational energy and to control the
input from the road that is transmitted to the vehicle [1].
Increasing in spring stiffness is not an option as its specification is governed by other vehicle
performance such as handling, stability etc. Hence, a detailed study has been performed to
understand the governing design variables. Damper and air spring “x” and “y” attachment
points have been selected as a design variables. Before ride analysis, full vehicle has been
modelled using Adams/Car and model has been validated by physical test. Simulation results
and physical test results has been shared as comparative.
*Corresponding author: Address: Şekerpınar Mah.Otomotiv Cad.Anadolu Isuzu Otomotiv A.Ş Çayırova/Kocaeli Kocaeli
41100 Turkey. E-mail address: [email protected], Phone: +902120878798
E.SERT et al./ ISITES2014 Karabuk - TURKEY
2. Correlation of the Adams/Car Model
When vehicle suspension is modelled, some requirements listed below have been
considered.
 System structural compenents (links, arms, hub, knuckle etc.) are represented as
rigid elements.
 Some of the inertias and masses that are unknown have been selected from Adams
library.
Figure 1. Adams/Car Model of the Portal Axle
Experimental correlation is essential in the development of analytical simulation models. A
methodology applied for correlating a bus model. Crossing over a known dimensions of bump
is one of the correlation tests. Therefore vehicle is crossed over the bump for physical test and
Adams/Car simulation. Test Equipments have been listed below.




GPS datalogger (VBOX)
Accelerometers
Displacement Transducers
Datalogger with video recorder (Video VBOX)
Figure 2. Vehicle Passed Over the Bump
Physical test and Adams/Car simulation results have been illustrated below as a comparative.
According to the results, Adams/Car vehicle model is definitely act as a real vehicle.
184
E.SERT et al./ ISITES2014 Karabuk - TURKEY
Figure 3. Left Rear Axle "z" Acceleration Value
Figure 5. Left Rear Axle "z" Displacement Value
Figure 4. Right Rear Axle "z" Acceleration Value
Figure 6. Right Rear Axle "z" Displacement Value
3. Adams/Car Sensitivity Analysis
Kinematic and Compliance (K&C) characteristics are the critical attribution to evaluate
suspension performance. Therefore, the wheel parameters of the vehicle which affect
vehicle’s ride performances were simulated and analyzed [2]. It is necessary to be reduced
bounce motion. Therefore, wheel vertical displacement has been selected design output. Full
vehicle model has been run under different roads. Then, rear left and right wheel vertical
displacement which are depend on the time have been shared below. Simulation result has
given the same result with full vehicle analysis.
Figure 7. Left Wheel Center Displacement Value
Figure 8. Right Wheel Center Displacement Value
A wheel-travel analysis allows to look at how the characteristics of a suspension change
throughout the vertical range of motion of the suspension.
Figure 9. Adams/Car Suspension Analysis
The upper limit of wheel-center displacement relative to the input position is 58 (mm) and the
lower limit of wheel-center displacement relative to the input position is -28 (mm).
185
E.SERT et al./ ISITES2014 Karabuk - TURKEY
Figure 10. Wheel Center Bump and Rebound Values
As seen on the Figure-10 magnitude of the compression and de-compression of the axle while
testing has similar results with the Adams/Car simulation. Coordinates of the displacement
transducer is transferred to Adams/Car model. The magnitude of the displacement value on
the this coordinate is shown on the Figure-11 and Figure-12
Figure 11. Left Wheel Center Displacement Value
In the simulation, the value of the displacement on the left rear axle is 36,92mm.
Figure 12. Right Wheel Center Displacement Value
Again in the simulation, the value of the displacement on the right rear axle is 23,21mm.
4. Parameter Optimization
Most used method is for decreasing the compression, to increase the stiffness of the air spring
or compression ratio of the damper to solve the problem. On the other hand increasing this
parameters has negative effect on the driving comfort [3]. For this reason we worked on
another method. Mounting of the air spring and damper coordinates were changed step by
step between “-40mm” and “+40mm” on the x and y axis. To have minumum displacement,
new coordinates have been optimised. For this optimisation, we used Adams/Insight module
and performed 256 steps.
4.1. Coordinate change on the “x” axis for Damper
Rows of the optimisation table is described below:
f_34 : Left rear damper (hpl_shock_to_frame)
f_44 : Left front damper (hpl_shock_to_frame_2)
f_35 : Right rear damper (hpr_shock_to_frame)
f_45 : Right front damper (hpr_shock_to_frame_2)
Minumum and maximum values of the x coordinates of the damper mount and left axle
displacement is shown in the Table 1 below:
186
E.SERT et al./ ISITES2014 Karabuk - TURKEY
Table 1. Minumum and Maximum "x" coordinates for Damper Mount
The coordinates calculated to have minumum displacement values of the axle are
Left rear damper (f_34.x) : 6622.5 mm
Left front damper (f_44.x) : 5157.5 mm
Right rear damper (f_35.x) : 6595.8 mm
Right front damper (f_45.x) : 5130.8 mm
These factors are varied one by one keeping other factors constant[7]. Figure 13 shows a set
of graphs for current experiment.
We can say that the left front, right rear and right front dampers show an irregular
distribution. Displacement value has been increased after declining trend. We can say that
left rear and left front dampers attachment points are very important and the most effective
parameters, whereas right rear and right front dampers attachment points are slightly less
important.
Figure 13. Effects of the Damper "x" Parameters
4.2. Coordinat change on the “y” axis for Damper
Minumum and maximum values of the y coordinates of the damper mount and left axle
displacement is shown in the Table 2 below:
187
E.SERT et al./ ISITES2014 Karabuk - TURKEY
Table 2. Minumum and Maximum "y" coordinates for Damper Mount
The coordinates calculated to have minumum displacement values of the axle are:
Left rear damper (f_34.y) : -942 mm
Left front damper (f_44.y) : -1022 mm
Right rear damper (f_35.y) : 1022 mm
Right front damper (f_45.y) : 1022 mm
These factors are varied one by one keeping other factors constant. Figure 14 shows a set of
graphs for current experiment.
We can say that the change of the y coordinates of the left front damper is increasing
displacement value. We can say that left rear and right front dampers attachment points are
very important and the most effective parameters, whereas right rear damper attachment point
is slightly less important.
Figure 14. Effects of the Damper "y" Parameters
4.3. Coordinat change on the “y” axis for Air Spring
Rows of the optimisation table is described below:
188
E.SERT et al./ ISITES2014 Karabuk - TURKEY
f_32 :Left Rear Air Spring (hpl_upper_spring_mount)
f_24 :Left Front Air Spring (hpl_upper_spring_mount_2)
f_33 :Right Rear Air Spring (hpr_upper_spring_mount)
f_25 :Right Front Air Spring (hpr_upper_spring_mount_2)
Minumum and maximum values of the y coordinates of the air spring mount and left axle
displacement is shown in the Table 3 below:
Table 3. Minumum and Maximum "y" coordinates for Air Spring Mount
The air spring coordinates calculated to have minumum displacement values of the axle are:
Left Rear Air Spring (f_32.y) : -733 mm
Left Front Air Spring (f_24.y) : -759.66 mm
Right Rear Air Spring (f_33.y) : 733 mm
Right Front Air Spring (f_25.y) : 733 mm
These factors are varied one by one keeping other factors constant. Figure 15 shows a set of
graphs for current experiment.
We can say that the left rear, right rear and right front dampers show an irregular distribution.
Displacement value has decrease after upward trend. We can say that left rear and right rear
dampers attachment points are very important and the most effective parameters.
189
E.SERT et al./ ISITES2014 Karabuk - TURKEY
Figure 15. Effects of the Air Spring "y" Parameters
4.4. Coordinat change on the “x” axis for Air Spring
Minumum and maximum values of the x coordinates of the air spring mount and left axle
displacement is shown in the Table 4 below:
Table 4.Minumum and Maximum "x" coordinates for Air Spring Mount
The air spring coordinates calculated to have minumum displacement values of the axle are:
Left Rear Air Spring (f_32.x) : 6635 mm
Left Front Air Spring (f_24.x) : 5091.67 mm
Right Rear Air Spring (f_33.x) : 6581.67 mm
Right Front Air Spring (f_25.x) : 5091.67 mm
These factors are varied one by one keeping other factors constant. Figure 16 shows a set of
graphs for current experiment.
We can say that the all of the parameters show an irregular distribution. Displacement value
has increased after declining trend. We can say that left rear and left front dampers
attachment points are very important and the most effective parameters, whereas right front
damper attachment point is slightly less important.
190
E.SERT et al./ ISITES2014 Karabuk - TURKEY
Figure 16. Effects of the Air Spring "x" Parameters
5. Results
In conclusion, change on the x and y mounting coordinates of the air spring and damper
optimised and effect of the parameters shown in this study. The most effective method to
increase driving comfort is to change y coordinate of the air spring. Changing of the damper
coordinates on the x axis and y axis are slightly less important. The most important two
parameters have been shown in figure 17, 18, 19 and 20 as to interact with one another [7].
Levels of factors are represented as three separate lines. Figure 17 shows interaction between
factors left rear and left front air spring "x" attachment points. Legend of plot displays the
levels of factor left rear air spring point. The levels of factor ‘left rear air spring’ are plotted as
three separate lines. The horizontal scale (X axis) shows the levels of the second factor 'left
front air spring’. The vertical scale (Y axis) is in the units of response displacement value.
Figure 17. Interaction between factors left rear and left front air spring "x" attachment points
Levels of factors are represented as three separate lines. Figure 18 shows interaction between
191
E.SERT et al./ ISITES2014 Karabuk - TURKEY
factors left rear and right rear air spring "y" attachment points. Legend of plot displays the
levels of factor left rear air spring point. The levels of factor ‘left rear air spring’ are plotted as
three separate lines. The horizontal scale (X axis) shows the levels of the second factor right
rear air spring’. The vertical scale (Y axis) is in the units of response displacement value.
Figure 18. Interaction between factors left rear and right front air spring "y" attachment points
Levels of factors are represented as three separate lines. Figure 19 shows interaction between
factors left rear and left front damper "x" attachment points. Legend of plot displays the levels
of factor left rear damper point. The levels of factor ‘left rear damper' are plotted as three
separate lines. The horizontal scale (X axis) shows the levels of the second factor 'left front
damper’. The vertical scale (Y axis) is in the units of response displacement value.
Figure 19. Interaction between factors left rear and left front damper "x" attachment points
Levels of factors are represented as three separate lines. Figure 20 shows interaction between
factors left rear and right front damper "y" attachment points. Legend of plot displays the
levels of factor left rear damper point. The levels of factor ‘left rear damper' are plotted as
three separate lines. The horizontal scale (X axis) shows the levels of the second factor 'right
front damper’. The vertical scale (Y axis) is in the units of response displacement value.
192
E.SERT et al./ ISITES2014 Karabuk - TURKEY
Figure 20. Interaction between factors left rear and right front damper "y" attachment points
References
[1] Xiaobin Ning, Cuiling Zhao, Jisheng Shen "Dynamic Analysis of Car Suspension Using
ADAMS/Car for Development of a Software Interface for Optimization"
[2] Xiang Meng, Wei Chen, Huihui Xue,Using Chassis Tuning Technology in Chassis Design
of Car, Springer-Verlag Berlin Heidelberg 2013
[3] Plackett, R.L., Burman, J.P.”The Design of Optimum Multifactorial Experiments.” In:
Biometrika 33 (1946)
[4] Ooi Jong Boon, “Analysis and Optimization of Portal Axle Unit Using Finite Element
Modelling and Simulation".2013
[5] Sohoni, V.N., Haug, E.J.”A State Space Technique for Optimal Design of Mechanisms”.
In: ASME Journal ofMechanical Design 104 (1982)
[6] ADAMS/Insight Training Guide,Release 2003
[7] E.Sert, Ağır Taşıtlarda Devrilme Dinamiğinin Eniyilenmesi ve Kontrol Sistemi Tasarımı,
İTÜ, 2014
[8]E. Sert, T. Çalışkan, H. Atasoy. “Zırhlı Aracın Statik Devrilme Karakteristiği” (Static
Rollover Characteristics of the Armored Vehicle), OTEKON 2014, Bursa.
193

Benzer belgeler